loading and running an aggregation in mongo shell

Question

I've coded an aggregation I want to run on a collection in the mongo shell. If I paste it directly into the shell, I have to do it line by line which is tedious and slow, especially when it doesn't work and I have to do it over. Therefore, I put it in a .js document and loaded it like this load("myaggregation.js") but that only returns true if the code is valid. How do I run the aggregation after loading it? I can't find any documentation about this. Do I have to use the nodejs driver?

Answer 1

Put your aggregation code in a function:

function my_aggregate() { 
    return db.foo.aggregate( [ ... ] );
}

and store that in your .js file.

Then run load to load it. You can also pass the filename on the commandline using the --shell commandline flag which will start the interactive shell after running any specified .js files.

Once the file is run your new function will be available to execute. Simply type

my_aggregate()

Update You do have to use an explicit return which I failed to mention. So in your case you would want something like:

function my_aggregate() {
    return db.zips.aggregate([{ $match: { state: { $in: [ "CA", "NY" ] } }},{ $group:{ _id : {  "state" : "$state","city" : "$city" }, pop : { $sum: "$pop"}}},{ $match: {  pop: { $gt: 25000 } }},{$group:{ _id : {   "city" : "$city"}, pop : { $avg: "$pop"}}}]);
}

Answer 2

Have you seen http://docs.mongodb.org/manual/tutorial/write-scripts-for-the-mongo-shell/?

Something like: mongo localhost:27017/test myjsfile.js

It also says that "You can execute a .js file from within the mongo shell, using the load() function".